SlideShare uma empresa Scribd logo
1 de 21
Baixar para ler offline
GO DOG GO
                         Robert Carr <racarr@gnome.org>

Friday, April 15, 2011
GO DOG GO
                         Robert Carr <racarr@gnome.org>

Friday, April 15, 2011
GO?
Friday, April 15, 2011
GOBJECT-DETECTOR


    • Go            board recognition library.

    • Rewritten             in C++ for Android NDK.

    • Some               issues with OpenCv on Android...




Friday, April 15, 2011
IMPROVED GRID DETECTION


    • Board    has dots to indicate
         “Hoshi” points (For example
         Tengen at 10,10)

    • Adjust   grid to fit better to
         Hoshi points.



Friday, April 15, 2011
MORE?



    • Score     boards. Requires a simple MonteCarlo AI or user
         interaction.

    • Live               broadcasting to KGS.




Friday, April 15, 2011
GIFU!


    • iTunes              for your Go game records.

    • Fully              queryable sqlite backed database.

    • Tested               with collection of 50,000 games.




Friday, April 15, 2011
TAGS AND COMMENTS


    • With     a collection of 50,000 games it can be difficult to
         remember what you are studying...or which game had a
         particular move.

    • Would       also be nice to comment game without disturbing the
         original file.



Friday, April 15, 2011
SGF PRINTER



    • Would     also be nice to print
         records in Kifu format.




Friday, April 15, 2011
GO AI



    • Really             hard :(




Friday, April 15, 2011
BOARD REPRESENTATION

    • Keep      track of blocks/
         liberties.

    • Zobrist            hashing

    • Positional          Super Ko

    • Symmetry             Checking.


Friday, April 15, 2011
AI ARCHITECTURE



    • Multiple            modules generate candidate moves.

    • Moves              are evaluated by Monte Carlo evaluator.




Friday, April 15, 2011
MONTE CARLO EVALUATOR


    • Core               of AI move evaluation.

    • Randomly     plays out thousands of games and calculates win
         percentage.

    • Playouts             can be “light” or “heavy”



Friday, April 15, 2011
PATTERN MATCHING



    • Match              subset of board against pattern.

    • Recognize              patterns independent of transposition, rotation,
         etc...




Friday, April 15, 2011
FUSEKI DATABASE


    • Compiled     from collection of
         pro and high dan amateur
         games.

    • Typically          relevant for first 50
         moves.



Friday, April 15, 2011
JOSEKI DATABASE


    • Patterns    in local corner
         positions.

    • Best    moves found for both
         sides.




Friday, April 15, 2011
CONNECTIVITY


    • Higher              level concept than
         blocks.

    • The    red stones are
         “connected”

    • Every              shape has weaknesses.



Friday, April 15, 2011
PROVERBIAL KNOWLEDGE


    • Black   would like to play
         inbetween the white stones
         on top.

    • Proverbial    knowledg says to
         build thickness first....pattern
         database has a suggestion!



Friday, April 15, 2011
THE PEEP
Friday, April 15, 2011
STRATEGIC DECISIONS

    • Proverbial   knowledge database will enable guiding selection of
         moves for MonteCarlo based on moves generated by various
         strategy modules.

    • In    previous example, “invasion” module, would query the
         “thickness” module for setup plays.

    • Such               plays are given heavier play outs in MonteCarlo.


Friday, April 15, 2011
QUESTIONS?




Friday, April 15, 2011

Mais conteúdo relacionado

Destaque

Rpi planner geoffrey_wright_7-22-2011
Rpi planner geoffrey_wright_7-22-2011Rpi planner geoffrey_wright_7-22-2011
Rpi planner geoffrey_wright_7-22-2011
mskmoorthy
 
Dynamorio rpioss-aug2011
Dynamorio rpioss-aug2011Dynamorio rpioss-aug2011
Dynamorio rpioss-aug2011
mskmoorthy
 
Mobile notifierspringsecondpresentation
Mobile notifierspringsecondpresentationMobile notifierspringsecondpresentation
Mobile notifierspringsecondpresentation
mskmoorthy
 
Concert spring 2011_presentation_2
Concert spring 2011_presentation_2Concert spring 2011_presentation_2
Concert spring 2011_presentation_2
mskmoorthy
 
Observatory 7 15-11
Observatory 7 15-11Observatory 7 15-11
Observatory 7 15-11
mskmoorthy
 

Destaque (18)

Rcos midterm
Rcos midtermRcos midterm
Rcos midterm
 
Nexus presentation1
Nexus presentation1Nexus presentation1
Nexus presentation1
 
Rpi planner geoffrey_wright_7-22-2011
Rpi planner geoffrey_wright_7-22-2011Rpi planner geoffrey_wright_7-22-2011
Rpi planner geoffrey_wright_7-22-2011
 
Dynamorio rpioss-aug2011
Dynamorio rpioss-aug2011Dynamorio rpioss-aug2011
Dynamorio rpioss-aug2011
 
Mobile notifierspringsecondpresentation
Mobile notifierspringsecondpresentationMobile notifierspringsecondpresentation
Mobile notifierspringsecondpresentation
 
Concert spring 2011_presentation_2
Concert spring 2011_presentation_2Concert spring 2011_presentation_2
Concert spring 2011_presentation_2
 
Flick slides
Flick slidesFlick slides
Flick slides
 
Crowdsourcing for geoint-11.11.11
Crowdsourcing for geoint-11.11.11Crowdsourcing for geoint-11.11.11
Crowdsourcing for geoint-11.11.11
 
Pres 1 2
Pres 1 2Pres 1 2
Pres 1 2
 
Observatory 7 15-11
Observatory 7 15-11Observatory 7 15-11
Observatory 7 15-11
 
Py chess learning-module
Py chess learning-modulePy chess learning-module
Py chess learning-module
 
Idea management systems drive competitiveness
Idea management systems drive competitivenessIdea management systems drive competitiveness
Idea management systems drive competitiveness
 
Idea Management Tools and Cases
Idea Management Tools and CasesIdea Management Tools and Cases
Idea Management Tools and Cases
 
5 lessons in idea management from the mouse
5 lessons in idea management from the mouse5 lessons in idea management from the mouse
5 lessons in idea management from the mouse
 
WLA 2015 Innovative Leadership Bootcamp: Idea Management
WLA 2015 Innovative Leadership Bootcamp: Idea ManagementWLA 2015 Innovative Leadership Bootcamp: Idea Management
WLA 2015 Innovative Leadership Bootcamp: Idea Management
 
Why ibx
Why ibx Why ibx
Why ibx
 
WALE 2014 Idea Management for Purposeful Innovation
WALE 2014 Idea Management for Purposeful InnovationWALE 2014 Idea Management for Purposeful Innovation
WALE 2014 Idea Management for Purposeful Innovation
 
What is Idea Management?
What is Idea Management?What is Idea Management?
What is Idea Management?
 

Semelhante a Godoggo

Milkyway@home rcos presentation_4_8_2011
Milkyway@home rcos presentation_4_8_2011Milkyway@home rcos presentation_4_8_2011
Milkyway@home rcos presentation_4_8_2011
mskmoorthy
 
Devopsdays Goteborg 2011 - State of the Union
Devopsdays Goteborg 2011 - State of the UnionDevopsdays Goteborg 2011 - State of the Union
Devopsdays Goteborg 2011 - State of the Union
John Willis
 
Monitoring is easy, why are we so bad at it presentation
Monitoring is easy, why are we so bad at it  presentationMonitoring is easy, why are we so bad at it  presentation
Monitoring is easy, why are we so bad at it presentation
Theo Schlossnagle
 
Clouds against the Floods (RubyConfBR2011)
Clouds against the Floods (RubyConfBR2011) Clouds against the Floods (RubyConfBR2011)
Clouds against the Floods (RubyConfBR2011)
Leonardo Borges
 
SVA Class 1
SVA Class 1SVA Class 1
SVA Class 1
boltron
 
Oop design magma rails 2011
Oop design   magma rails 2011Oop design   magma rails 2011
Oop design magma rails 2011
MagmaConf
 

Semelhante a Godoggo (20)

Milkyway@home rcos presentation_4_8_2011
Milkyway@home rcos presentation_4_8_2011Milkyway@home rcos presentation_4_8_2011
Milkyway@home rcos presentation_4_8_2011
 
Devopsdays Goteborg 2011 - State of the Union
Devopsdays Goteborg 2011 - State of the UnionDevopsdays Goteborg 2011 - State of the Union
Devopsdays Goteborg 2011 - State of the Union
 
Atlassian RoadTrip 2011 Slide Deck
Atlassian RoadTrip 2011 Slide DeckAtlassian RoadTrip 2011 Slide Deck
Atlassian RoadTrip 2011 Slide Deck
 
Discussing Java's Future
Discussing Java's FutureDiscussing Java's Future
Discussing Java's Future
 
Monitoring is easy, why are we so bad at it presentation
Monitoring is easy, why are we so bad at it  presentationMonitoring is easy, why are we so bad at it  presentation
Monitoring is easy, why are we so bad at it presentation
 
Fast Map Interaction without Flash
Fast Map Interaction without FlashFast Map Interaction without Flash
Fast Map Interaction without Flash
 
Webops dashboards
Webops dashboardsWebops dashboards
Webops dashboards
 
Puppet camp europe 2011 hackability
Puppet camp europe 2011   hackabilityPuppet camp europe 2011   hackability
Puppet camp europe 2011 hackability
 
Ignite@DevOpsDays - Why devs need ops
Ignite@DevOpsDays - Why devs need opsIgnite@DevOpsDays - Why devs need ops
Ignite@DevOpsDays - Why devs need ops
 
High quality iOS development
High quality iOS developmentHigh quality iOS development
High quality iOS development
 
Bonfire... How'd You Do That?! - AtlasCamp 2011
Bonfire... How'd You Do That?! - AtlasCamp 2011Bonfire... How'd You Do That?! - AtlasCamp 2011
Bonfire... How'd You Do That?! - AtlasCamp 2011
 
Clouds against the Floods (RubyConfBR2011)
Clouds against the Floods (RubyConfBR2011) Clouds against the Floods (RubyConfBR2011)
Clouds against the Floods (RubyConfBR2011)
 
SVA Class 1
SVA Class 1SVA Class 1
SVA Class 1
 
Oop design magma rails 2011
Oop design   magma rails 2011Oop design   magma rails 2011
Oop design magma rails 2011
 
obs-tdd-intro
obs-tdd-introobs-tdd-intro
obs-tdd-intro
 
A new world with cdi en
A new world with cdi enA new world with cdi en
A new world with cdi en
 
The Hitchhiker's guide to mobile development
The Hitchhiker's guide to mobile developmentThe Hitchhiker's guide to mobile development
The Hitchhiker's guide to mobile development
 
Gtmf2011 2011.06.07 slideshare
Gtmf2011 2011.06.07 slideshareGtmf2011 2011.06.07 slideshare
Gtmf2011 2011.06.07 slideshare
 
CTF: Bringing back more than sexy!
CTF: Bringing back more than sexy!CTF: Bringing back more than sexy!
CTF: Bringing back more than sexy!
 
Eurobot-OHW
Eurobot-OHWEurobot-OHW
Eurobot-OHW
 

Mais de mskmoorthy (20)

Shuttle trackersecondpresentationfall2011
Shuttle trackersecondpresentationfall2011Shuttle trackersecondpresentationfall2011
Shuttle trackersecondpresentationfall2011
 
Rcos intro-2
Rcos intro-2Rcos intro-2
Rcos intro-2
 
Mobile shuttle tracker_fall_2011_first_present
Mobile shuttle tracker_fall_2011_first_presentMobile shuttle tracker_fall_2011_first_present
Mobile shuttle tracker_fall_2011_first_present
 
Rcos presentation 9-23-2011
Rcos presentation 9-23-2011Rcos presentation 9-23-2011
Rcos presentation 9-23-2011
 
Rcos presentation
Rcos presentationRcos presentation
Rcos presentation
 
Auto scheduler presentation_2
Auto scheduler presentation_2Auto scheduler presentation_2
Auto scheduler presentation_2
 
Second presentation idea_bank
Second presentation idea_bankSecond presentation idea_bank
Second presentation idea_bank
 
Scrutiny 2
Scrutiny 2Scrutiny 2
Scrutiny 2
 
Rpi planner geoffrey_wright_7-22-2011
Rpi planner geoffrey_wright_7-22-2011Rpi planner geoffrey_wright_7-22-2011
Rpi planner geoffrey_wright_7-22-2011
 
Sean austin uir-2
Sean austin uir-2Sean austin uir-2
Sean austin uir-2
 
Nexus2 7-22-1011
Nexus2 7-22-1011Nexus2 7-22-1011
Nexus2 7-22-1011
 
Flagship safety, hfoss-7-22-2011
Flagship safety, hfoss-7-22-2011Flagship safety, hfoss-7-22-2011
Flagship safety, hfoss-7-22-2011
 
Olympus pesentation2
Olympus pesentation2Olympus pesentation2
Olympus pesentation2
 
8.7.2011 agml
8.7.2011 agml8.7.2011 agml
8.7.2011 agml
 
Concerto conmoto
Concerto conmotoConcerto conmoto
Concerto conmoto
 
Koala pres1
Koala pres1Koala pres1
Koala pres1
 
Scrutiny 07-01-2011.pdf
Scrutiny 07-01-2011.pdfScrutiny 07-01-2011.pdf
Scrutiny 07-01-2011.pdf
 
Koala pres1
Koala pres1Koala pres1
Koala pres1
 
Initial presentation
Initial presentationInitial presentation
Initial presentation
 
Flagship safety
Flagship safetyFlagship safety
Flagship safety
 

Último

Gardella_PRCampaignConclusion Pitch Letter
Gardella_PRCampaignConclusion Pitch LetterGardella_PRCampaignConclusion Pitch Letter
Gardella_PRCampaignConclusion Pitch Letter
MateoGardella
 
1029 - Danh muc Sach Giao Khoa 10 . pdf
1029 -  Danh muc Sach Giao Khoa 10 . pdf1029 -  Danh muc Sach Giao Khoa 10 . pdf
1029 - Danh muc Sach Giao Khoa 10 . pdf
QucHHunhnh
 
Activity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfActivity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdf
ciinovamais
 
1029-Danh muc Sach Giao Khoa khoi 6.pdf
1029-Danh muc Sach Giao Khoa khoi  6.pdf1029-Danh muc Sach Giao Khoa khoi  6.pdf
1029-Danh muc Sach Giao Khoa khoi 6.pdf
QucHHunhnh
 
The basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptxThe basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptx
heathfieldcps1
 

Último (20)

APM Welcome, APM North West Network Conference, Synergies Across Sectors
APM Welcome, APM North West Network Conference, Synergies Across SectorsAPM Welcome, APM North West Network Conference, Synergies Across Sectors
APM Welcome, APM North West Network Conference, Synergies Across Sectors
 
Gardella_PRCampaignConclusion Pitch Letter
Gardella_PRCampaignConclusion Pitch LetterGardella_PRCampaignConclusion Pitch Letter
Gardella_PRCampaignConclusion Pitch Letter
 
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
 
SECOND SEMESTER TOPIC COVERAGE SY 2023-2024 Trends, Networks, and Critical Th...
SECOND SEMESTER TOPIC COVERAGE SY 2023-2024 Trends, Networks, and Critical Th...SECOND SEMESTER TOPIC COVERAGE SY 2023-2024 Trends, Networks, and Critical Th...
SECOND SEMESTER TOPIC COVERAGE SY 2023-2024 Trends, Networks, and Critical Th...
 
Application orientated numerical on hev.ppt
Application orientated numerical on hev.pptApplication orientated numerical on hev.ppt
Application orientated numerical on hev.ppt
 
Unit-IV- Pharma. Marketing Channels.pptx
Unit-IV- Pharma. Marketing Channels.pptxUnit-IV- Pharma. Marketing Channels.pptx
Unit-IV- Pharma. Marketing Channels.pptx
 
1029 - Danh muc Sach Giao Khoa 10 . pdf
1029 -  Danh muc Sach Giao Khoa 10 . pdf1029 -  Danh muc Sach Giao Khoa 10 . pdf
1029 - Danh muc Sach Giao Khoa 10 . pdf
 
Class 11th Physics NEET formula sheet pdf
Class 11th Physics NEET formula sheet pdfClass 11th Physics NEET formula sheet pdf
Class 11th Physics NEET formula sheet pdf
 
Activity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfActivity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdf
 
Paris 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activityParis 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activity
 
Introduction to Nonprofit Accounting: The Basics
Introduction to Nonprofit Accounting: The BasicsIntroduction to Nonprofit Accounting: The Basics
Introduction to Nonprofit Accounting: The Basics
 
Ecological Succession. ( ECOSYSTEM, B. Pharmacy, 1st Year, Sem-II, Environmen...
Ecological Succession. ( ECOSYSTEM, B. Pharmacy, 1st Year, Sem-II, Environmen...Ecological Succession. ( ECOSYSTEM, B. Pharmacy, 1st Year, Sem-II, Environmen...
Ecological Succession. ( ECOSYSTEM, B. Pharmacy, 1st Year, Sem-II, Environmen...
 
Unit-V; Pricing (Pharma Marketing Management).pptx
Unit-V; Pricing (Pharma Marketing Management).pptxUnit-V; Pricing (Pharma Marketing Management).pptx
Unit-V; Pricing (Pharma Marketing Management).pptx
 
1029-Danh muc Sach Giao Khoa khoi 6.pdf
1029-Danh muc Sach Giao Khoa khoi  6.pdf1029-Danh muc Sach Giao Khoa khoi  6.pdf
1029-Danh muc Sach Giao Khoa khoi 6.pdf
 
Sports & Fitness Value Added Course FY..
Sports & Fitness Value Added Course FY..Sports & Fitness Value Added Course FY..
Sports & Fitness Value Added Course FY..
 
The basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptxThe basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptx
 
microwave assisted reaction. General introduction
microwave assisted reaction. General introductionmicrowave assisted reaction. General introduction
microwave assisted reaction. General introduction
 
Basic Civil Engineering first year Notes- Chapter 4 Building.pptx
Basic Civil Engineering first year Notes- Chapter 4 Building.pptxBasic Civil Engineering first year Notes- Chapter 4 Building.pptx
Basic Civil Engineering first year Notes- Chapter 4 Building.pptx
 
This PowerPoint helps students to consider the concept of infinity.
This PowerPoint helps students to consider the concept of infinity.This PowerPoint helps students to consider the concept of infinity.
This PowerPoint helps students to consider the concept of infinity.
 
Mattingly "AI & Prompt Design: Structured Data, Assistants, & RAG"
Mattingly "AI & Prompt Design: Structured Data, Assistants, & RAG"Mattingly "AI & Prompt Design: Structured Data, Assistants, & RAG"
Mattingly "AI & Prompt Design: Structured Data, Assistants, & RAG"
 

Godoggo

  • 1. GO DOG GO Robert Carr <racarr@gnome.org> Friday, April 15, 2011
  • 2. GO DOG GO Robert Carr <racarr@gnome.org> Friday, April 15, 2011
  • 4. GOBJECT-DETECTOR • Go board recognition library. • Rewritten in C++ for Android NDK. • Some issues with OpenCv on Android... Friday, April 15, 2011
  • 5. IMPROVED GRID DETECTION • Board has dots to indicate “Hoshi” points (For example Tengen at 10,10) • Adjust grid to fit better to Hoshi points. Friday, April 15, 2011
  • 6. MORE? • Score boards. Requires a simple MonteCarlo AI or user interaction. • Live broadcasting to KGS. Friday, April 15, 2011
  • 7. GIFU! • iTunes for your Go game records. • Fully queryable sqlite backed database. • Tested with collection of 50,000 games. Friday, April 15, 2011
  • 8. TAGS AND COMMENTS • With a collection of 50,000 games it can be difficult to remember what you are studying...or which game had a particular move. • Would also be nice to comment game without disturbing the original file. Friday, April 15, 2011
  • 9. SGF PRINTER • Would also be nice to print records in Kifu format. Friday, April 15, 2011
  • 10. GO AI • Really hard :( Friday, April 15, 2011
  • 11. BOARD REPRESENTATION • Keep track of blocks/ liberties. • Zobrist hashing • Positional Super Ko • Symmetry Checking. Friday, April 15, 2011
  • 12. AI ARCHITECTURE • Multiple modules generate candidate moves. • Moves are evaluated by Monte Carlo evaluator. Friday, April 15, 2011
  • 13. MONTE CARLO EVALUATOR • Core of AI move evaluation. • Randomly plays out thousands of games and calculates win percentage. • Playouts can be “light” or “heavy” Friday, April 15, 2011
  • 14. PATTERN MATCHING • Match subset of board against pattern. • Recognize patterns independent of transposition, rotation, etc... Friday, April 15, 2011
  • 15. FUSEKI DATABASE • Compiled from collection of pro and high dan amateur games. • Typically relevant for first 50 moves. Friday, April 15, 2011
  • 16. JOSEKI DATABASE • Patterns in local corner positions. • Best moves found for both sides. Friday, April 15, 2011
  • 17. CONNECTIVITY • Higher level concept than blocks. • The red stones are “connected” • Every shape has weaknesses. Friday, April 15, 2011
  • 18. PROVERBIAL KNOWLEDGE • Black would like to play inbetween the white stones on top. • Proverbial knowledg says to build thickness first....pattern database has a suggestion! Friday, April 15, 2011
  • 20. STRATEGIC DECISIONS • Proverbial knowledge database will enable guiding selection of moves for MonteCarlo based on moves generated by various strategy modules. • In previous example, “invasion” module, would query the “thickness” module for setup plays. • Such plays are given heavier play outs in MonteCarlo. Friday, April 15, 2011